Skip to content

We're upgrading our operations to serve you better. Orders ship as usual from Laval, QC. Questions? Contact us

Bitcoin accepted at checkout  |  Ships from Laval, QC, Canada  |  Expert support since 2016

1325 Rue Bergar, Laval, QC support@d-central.tech 1-855-753-9997 Mon-Fri: 8:30 - 18:00
Current

RTX A5000

NVIDIA · dual-slot blower · Released April 2021

Dual-slot blower with 24 GB and ECC. The professional's 3090 — same VRAM, quieter, rack-ready.

Hardware spec sheet

VendorNVIDIA
CategoryGPU
VRAM / memory24 GB
Memory bandwidth768 GB/s
FP16 TFLOPS27.8
INT8 TOPS222
TDP230 W
ArchitectureAmpere
Form factordual-slot blower
Release dateApril 2021
Street price (USD)800-1100 (used)
120V note230W each; two A5000s on 120V/15A with a 1000W PSU is the practical limit.

The RTX A5000 is the dual-slot blower Ampere workstation card: 24 GB of ECC GDDR6, 768 GB/s, 230 W TDP. Launched alongside the A4000 in April 2021, it slots between the A4000 (16 GB, 140 W) and the A6000 (48 GB, 300 W). Ampere architecture identical to the 3090 — same CUDA core count even — but with ECC memory, a blower cooler, and NVIDIA’s professional driver stack.

Who it’s for: professionals, studios, and Hashcenter operators who want 3090-class performance in a rack-friendly package with ECC for workload reliability. Used prices \$800–1100 as of 2026.

Models it runs comfortably: identical parameter envelope to the RTX 3090 (Llama 3 70B at Q4, Qwen 2.5 32B at Q5_K_M), with slightly lower bandwidth (768 vs 936 GB/s) meaning ~20% slower tok/s on memory-bound workloads. The ECC and blower form factor usually justify the trade-off for production use.

Hashcenter notes: dual-slot blower — two A5000s fit in four slots and dissipate 460 W total through front-to-back airflow, ideal for a 4U chassis. Quieter than a 3090 blower under load because the lower clocks keep fan RPM down. 230 W each means two cards on 120V/15A with a 1000 W PSU. Credit NVIDIA’s Quadro/RTX Pro lineage for the engineering.

Further reading: This card is a core component of a pleb-grade AI Hashcenter. Pair it with the sovereignty argument in the Sovereign AI for Bitcoiners Manifesto, or look at how the same 120V envelope powers a Bitcoin space heater in our mining catalog. Running both workloads on one rig? See Heating Your Home With Inference.

Models that run on this hardware

Gemma 3 Gemma runs at Q5_K_M with headroom Gemma 2 Gemma runs at Q5_K_M with headroom Mistral Small 3 Mistral runs at Q5_K_M with headroom Phi-4 Phi runs at Q8 / FP16 comfortably FLUX.1 schnell FLUX runs at Q8 / FP16 comfortably FLUX.1 dev FLUX runs at Q8 / FP16 comfortably Stable Diffusion 3.5 Stable Diffusion runs at Q8 / FP16 comfortably Mistral 7B Mistral runs at Q8 / FP16 comfortably

Get it running

  1. 01 Install Ollama →

    Ten-minute local LLM runtime. One binary, zero cloud.

  2. 02 Give it a UI →

    Open-WebUI turns Ollama into a self-hosted ChatGPT.

  3. 03 Which runner? →

    LM Studio vs Ollama vs llama.cpp — pick the right runtime for your rig.

Further reading: Heating your home with inference for turning this card into a winter-heat source, and the Sovereign AI for Bitcoiners Manifesto for the bigger picture on owner-operated AI.